paper-with-me

Papers

Benchmarking Compositionality with Formal Languages

2022-08-17 · COLING 2022 10 · Josef Valvoda, Naomi Saphra, Jonathan Rawski, Adina Williams, Ryan Cotterell

Recombining known primitive concepts into larger novel combinations is a quintessentially human cognitive capability. Whether large neural models in NLP can acquire this ability while learning from data is an open question. In this paper, we investigate this problem from the perspective of formal languages. We use deterministic finite-state transducers to make an unbounded number of datasets with controllable properties governing compositionality. By randomly sampling over many transducers, we explore which of their properties contribute to learnability of a compositional relation by a neural network. We find that the models either learn the relations completely or not at all. The key is transition coverage, setting a soft learnability limit at 400 examples per transition.

📄 PDF Abstract BibTeX arXiv:2208.08195

Code (1)

valvoda/neuraltransducer 공식 구현 pytorch

Tasks

BenchmarkingOpen-Ended Question Answering

Similar Papers 제목 키워드 기반

The Combinatorics of \textit{Salva Veritate} Principles

2022-01-13 · Norman E. Trushaev

Various concepts of grammatical compositionality arise in many theories of both natural and artificial languages, and often play a key role in accounts of the syntax-semantics interface. We propose that many instances of…

On Using Distribution-Based Compositionality Assessment to Evaluate Compositional Generalisation in Machine Translation

2023-11-14 · Anssi Moisio, Mathias Creutz, Mikko Kurimo

Compositional generalisation (CG), in NLP and in machine learning more generally, has been assessed mostly using artificial datasets. It is important to develop benchmarks to assess CG also in real-world natural language…

BenchmarkingMachine TranslationNMT

Measuring Compositionality in Representation Learning

2019-02-19 · ICLR 2019 5 · Jacob Andreas

Many machine learning algorithms represent input data with vector embeddings or discrete codes. When inputs exhibit compositional structure (e.g. objects built from parts or procedures from subroutines), it is natural to…

BIG-bench Machine LearningRepresentation Learning

Internal and external pressures on language emergence: least effort, object constancy and frequency

2020-04-08 · Findings of the Association for Computational Linguistics 2020 · Diana Rodríguez Luna, Edoardo Maria Ponti, Dieuwke Hupkes, Elia Bruni

In previous work, artificial agents were shown to achieve almost perfect accuracy in referential games where they have to communicate to identify images. Nevertheless, the resulting communication protocols rarely display…

DiagnosticPosition

Internal and External Pressures on Language Emergence: Least Effort, Object Constancy and Frequency

2019-06-07 · Anonymous

In previous work, artificial agents were shown to achieve almost perfect accuracy in referential games where they have to communicate to identify images. Nevertheless, the resulting communication protocols rarely display…

DiagnosticPosition